Incremental discontinuous phrase structure parsing with the GAP transition

Abstract

This article introduces a novel transition system for discontinuous lexicalized constituent parsing called SR-GAP. It is an extension of the shift-reduce algorithm with an additional gap transition. Evaluation on two German treebanks shows that SR-GAP outperforms the previous best transitionbased discontinuous parser (Maier, 2015) by a large margin (it is notably twice as accurate on the prediction of discontinuous constituents), and is competitive with the state of the art (Ferńandez-Gonźalez and Martins, 2015). As a side contribution, we adapt span features (Hall et al., 2014) to discontinuous parsing.